Sysctl Sentinel Bloat Removal Wrapping Up In Linux 6.10

[phoronix]

Phoronix: Sysctl Sentinel Bloat Removal Wrapping Up In Linux 6.10

The year-long effort to removal the sysctl sentinel for clearing bloat from the kernel and allowing faster build times should be crossing the finish line in Linux 6.10...

https://www.phoronix.com/news/Sysctl-Finish-Line-Linux-6.10

[rene]

can't wait to unleash all the extra performance of my Sgi O2: https://www.youtube.com/watch?v=Gri9LJGhC2g

[varikonniemi]

so many articles about this issue, yet not once has it been elaborated wtf it is or why is it a big task like this spanning multiple years/releases?

The way it has been covered has anyone except kernel hackers scratching their head thinking "how hard can it be to remove curly braces in a few places"

[yoshi314]

i think it's the potential of breakage, since it touches so many places.

i recall how BKL was removed from the kernel, and it was a coordinated process that took a while.

[Kjell]

Isn't automatic linting supposed to take care of such things via a git hook?

[npwx]

So what Michael has been constantly describing as and linking to as "faster kernel build times" does in fact state "reduce the overall build time *size*", not build time itself. So yeah, it removes a hand full or two of null bytes. Not that I'm against that, but reading an incorrect description of some random unimporant patchset for the nth time now is starting to annoy me. So hoorah, its finally done.

[EphemeralEft]

So what Michael has been constantly describing as and linking to as "faster kernel build times" does in fact state "reduce the overall build time *size*", not build time itself. So yeah, it removes a hand full or two of null bytes. Not that I'm against that, but reading an incorrect description of some random unimporant patchset for the nth time now is starting to annoy me. So hoorah, its finally done.

Not every article can be “systemd-btrfs now required for Snap kernel”.